#14776: Add equality rule for `eqString` -------------------------------------+------------------------------------- Reporter: mpickering | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by carter): That’s no sound for floating point when nan is treated as a valid value rather than an error/ exception trigger. In the former case this optimization will break ALL isNan code in Haskell. If we also have a mode where calculations that would produce a nan throw an exception, this rule would be safe. Is there a way to provide this rule but prevent it from running on floats and doubles ? -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14776#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler